Zhiyuan Dai

Dr. Zhiyuan Dai

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 October 2021 in Railway Engineering Science
Verified
Journal Article
Open access
PDF file
ZDZhiyuan Dai
Zhiyuan Dai
Tian LiTian Li
Tian Li
NZNing Zhou
Ning Zhou
JZJiye Zhang
Jiye Zhang
WZWeihua Zhang
Weihua Zhang
Aiming at the problem that aerodynamic uplift forces of the pantograph running in the knuckle-downstream and knuckle-upstream conditions are inconsistent, and their magnitudes do not satisfy the corresponding standard, the aerodynamic uplift forces of pantographs with baffles are numerically investigated, and an optimization method to determine the baffle angle is proposed. First, the error between the aerodynamic resistances of the pantograph obtained by numerical simulation and wind tunnel test is less than 5%, which indicates the accuracy of the numerical simulation method. Second, the original pantograph and pantographs equipped with three different baffles are numerically simulated to obtain the aerodynamic forces and moments of the pantograph components. Three different angles for the baffles are −17°, 0° and 17°. Then the multibody simulation is used to calculate the aerodynamic uplift force of the pantograph, and the optimal range for the baffle angle is determined. Results show that the lift force of the baffle increases with the increment of the angle in the knuckle-downstream condition, whereas the lift force of the baffle decreases with the increment of the angle in the knuckle-upstream condition. According to the results of the aerodynamic uplift force, the optimal angle of the baffle is determined to be 4.75° when the running speed is 350 km/h, and pantograph–catenary contact forces are 128.89 N and 129.15 N under the knuckle-downstream and knuckle-upstream operating conditions, respectively, which are almost equal and both meet the requirements of the standard EN50367:2012.

10 February 2024 in Journal of Wind Engineering and Industrial Aerodynamics
Verified
Journal Article
Closed access
ZDZhiyuan Dai
Zhiyuan Dai
Tian LiTian Li
Tian Li
WZWeihua Zhang
Weihua Zhang
JZJiye Zhang
Jiye Zhang

1 March 2024 in Physics of Fluids
Verified
Journal Article
Open access
PDF file
ZDZhiyuan Dai
Zhiyuan Dai
Tian LiTian Li
Tian Li
SKSiniša Krajnović
Siniša Krajnović
WZWeihua Zhang
Weihua Zhang
For models with large numerical simulation costs, such as high-speed trains, using as few samples as possible to construct a high-precision surrogate model during aerodynamic multi-objective optimization is critical to improving optimization efficiency. This study proposes a sequential infill criterion (SIC) appropriate for the Kriging surrogate model to address this issue. Three multi-objective functions are employed to test the feasibility of constructing a surrogate model based on SIC, and the SIC surrogate model then performs multi-objective aerodynamic optimizations on the high-speed train. The findings indicate that the expected improvement infill criterion (EIC) in the first stage can enhance the global prediction accuracy of the SIC. An infill criterion based on EIC that fuses gradient information (PGEIC) in the second stage is proposed to seek samples in the Pareto front. The PGEIC surrogate model achieves the lowest generational distance and prediction error. The performance of EIC for global search, EIC for Pareto front search, and infill criterion for Pareto front search using only gradient information is poor. The final PGEIC–SIC surrogate model of train aerodynamics has less than 1% prediction error for the three optimization objectives. The optimal solution reduces the aerodynamic drag force of the head car and the aerodynamic drag and lift force of the tail car by 4.15%, 3.21%, and 3.56%, respectively, compared with the original model. Furthermore, sensitivity analysis of key parameters revealed that the nose height v1, cab window height v3, and lower contour line have a greater impact on aerodynamic forces. Moreover, the nose and cab window heights of the optimal model have been reduced, and the lower contour line is concave. Correspondingly, the streamlined shape appears more rounded and slender.

20 May 2022 in Journal of Wind Engineering and Industrial Aerodynamics
Verified
Journal Article
Closed access
MYMengge Yu
Mengge Yu
JLJiali Liu
Jiali Liu
QZQian Zhang
Qian Zhang
ZDZhiyuan Dai
Zhiyuan Dai

1 January 2024 in Journal of Mechanical Engineering
Verified
Journal Article
Closed access
PDF file
戴志戴志远,李田,张卫华,张继业
戴志远,李田,张卫华,张继业
ZDZhiyuan Dai
Zhiyuan Dai

1 September 2026 in Physics of Fluids
Verified
Journal Article
Closed access
YTYe Tan
Ye Tan
DQDeng Qin
Deng Qin
Tian LiTian Li
Tian Li
ZDZhiyuan Dai
Zhiyuan Dai
JZJiye Zhang
Jiye Zhang
Nose-shape optimization of high-speed trains requires repeated high-fidelity computational fluid dynamics (CFD) simulations for different candidate geometries. Conventional surrogate models mainly predict integrated aerodynamic coefficients and cannot reconstruct surface pressure distributions needed to investigate aerodynamic mechanisms. This study proposes a geometry-aware graph neural network incorporating axial-attention and pressure-gradient supervision for rapid prediction of surface pressure fields of high-speed trains with different streamlined nose lengths. Surface coordinates, normal vectors, and curvature represent local geometry. Axial attention is used to represent long-range interactions between streamwise regions, while pressure-gradient supervision improves local differential consistency. Flow field analysis shows that pressure and velocity vary continuously along the train, with geometry-induced changes concentrated near the streamlined nose and curvature transition regions. For the 10 m interpolation experiment, the model achieves a mean absolute error (MAE) of 57.87 Pa and a coefficient of determination of 0.9469. Compared with the baseline graph neural network, the MAE is reduced by 37.2%, while the coefficient of determination increases from 0.8193 to 0.9469. The predicted whole-train pressure-drag coefficient is 0.1311, compared with the CFD value of 0.1355, corresponding to a relative error of 3.32%. The predicted whole-train viscous-drag coefficient is 0.1073, compared with the CFD value of 0.1069, corresponding to a relative error of 0.37%; the errors for individual cars are below 0.5%. Error distribution analysis shows that the model shows lower pressure prediction errors in geometrically and aerodynamically complex regions.